Meta’s current Commercial Terms preview lists November 20, 2026, as the effective date for the revised terms. The clause authorizes automated software and “test credentials” to access and collect material from websites associated with users’ products or services. A report published October 23 said Meta had moved the date from October 30 to November 20.
What Meta’s current clause covers
The terms apply when people use Meta Company Products for business or commercial purposes, including advertising, marketing messages, product listings, managing Pages or business Groups, and using Ads Manager or Meta Business Suite. The crawling clause covers associated domains, pages and URLs, and allows tools to index, cache, analyze, retrieve or crawl material and periodically pull updates.
The clause lists providing, developing, improving, researching and maintaining advertising and business products, as well as maintaining the integrity and security of Meta products, as purposes for the collected content. It does not name AI training. Media and intellectual-property attorney Matt Topic has argued that “develop” could include training, an interpretation of the wording rather than a decision on this clause.
Meta has separately said website owners can use robots.txt to manage crawler access. The clause itself makes no express promise to honor robots.txt, registration walls or paywalls. For businesses that disagree with an update, Meta’s terms describe stopping business or commercial use of its products as the alternative. They do not state that blocking a crawler will trigger account suspension.
What publishers are asking Meta to change
On October 22, News/Media Alliance asked Meta CEO Mark Zuckerberg to withdraw the terms and discuss changes with publishers. The group also called for a separate opt-in to scraping, clarity about which products and bots are involved, and respect for robots.txt, registration walls and paywalls.
Four European publisher organizations, the European Magazine Media Association, European Newspaper Publishers’ Association, European Press Council and News Media Europe, asked the European Commission to block the terms in their current or equivalent form. Spain’s Association of Information Media (AMI) backed their initiative. These are requests by publisher groups, not regulatory decisions.
What Kadrey v. Meta decided
In 2025, a federal district court found fair use in Meta’s use of copyrighted books to train Llama on the specific record in Kadrey v. Meta. The U.S. Copyright Office’s summary describes that case-specific ruling. It did not decide whether the current commercial-terms clause is enforceable or settle every question about web scraping and AI training.
